I'll admit to not being particularly well-acquainted with VR, but a few observations:
1) This is being used as a target for Servo. Even if it ends up being useless (I won't pretend to be able to predict the future), the investigation into getting consistent 90+ FPS performance out of web rendering engines, regardless of content, is going to be excellent for the web. VR is a challenging environment for the Servo team to prove its technological prowess, and for many engineers, a challenging environment is exactly the kind of motivation necessary to push at boundaries.
2) Mozilla missed the "social" hype train and their mobile OS project was widely considered too little, too late (not to mention the terrible performance of mobile web engines at the time). Both of these areas are dominated by large, (effectively) proprietary vendors. Facebook's purchase of Oculus signaled to many that "social" was going to move to yet another platform. Mozilla's goal here is to get out in front of a potential VR hype train, keeping the open web as a major player in order to keep their mission relevant.
3) Mozilla believes the web is the ultimate platform for connecting people. It is one of the few markets with a political landmap that encourages collaboration between the multiple players. Therefore, for Mozilla, the capabilities of the web must always be expanded to prevent balkanization of the Internet into product-specific networks, such as Apple's iMessage. I am actually encouraged by their willingness to jump in front of VR, despite my own belief that VR won't be as relevant as many prognosticators think.
This is my analysis, take it with multiple grains of salt.